??? 07/11/07 18:51 Read: times |
#141778 - Here we go again ... Responding to: ???'s previous message |
Apparently you guys have decided to go "ad hominem" and start throwing out "red herrings" rather than solely addressing the O/P's original question.
It's true, he seems misguided. No amount of "beating up" on him will likely change his mind, thoguh. It may make him go away, but it won't make him change his mind. The fact is, it's true that using Jeff Post's code would be much wiser than reinventing the wheel. It's true that trying to use a keypad and LCD is a mite poorly thought out. It's true that trying to program a device without the aid of a PC, when those are so ubiquitous, seems poorly considered. Perhaps there are some details he hasn't yet revealed about his goals. Perhaps what should be recommended is a new approach to a simulator, wherein all peripherals, internal or external can be simulated by means of an external module linked at compile time. That way, the internal timing and peripheral event timing can be accurately simulated, right down to interrupt responses and randomized input stimuli. It should simulate clock-tick by clock-tick, providing precise simulation of timers, serial ports, ADC's, DAC's, reset events, FLASH writes, etc. It should also simulate the progress of code execution with each possible branch both taken and bypassed. The result should be an exhaustive list of entry points encountered, and a listing of all unexecuted code space. A disassembler won't do that. No simulator I've ever encountered will do that, either. It would be a MAJOR enhancement to the reverse-engineering toolset, however. This would be a major challenge, so it's no surprise to me that it hasn't been done yet. I, for one, have yet to learn to use Jeff Post's disassembler. ... Maybe someday ... RE |